1. Field
Example aspects of the present invention generally relate to media item recommendations, and more particularly to systems, methods, and computer program products for providing a recommendation of a media item.
2. Related Art
The digitization of music, movies, and games, as well as the improvement in electronic delivery techniques, have changed the way consumers experience such media content. Consumers can download digital music, movies, or games via the Internet with the click of a mouse, and can enjoy them at their convenience. Consumers can also have media content streamed from a Web site to a consumer device, such as a computer, a television, or portable device, upon demand.
In addition, the variety of media content available to consumers today is wider than ever. Consumers can browse vast collections of music, movies, and games via Internet or television browsers to identify a particular media item to download or stream. Locating desired media items within these collections, however, can be difficult given their enormity.
To avoid the need for consumers to browse these vast collections, content providers provide consumers with recommendations of media items. Recommending media items to consumers promotes the discovery of new media items and increases consumer satisfaction with the content provider.
One conventional approach to providing consumers with recommendations of media items has been to compute an explicit numerical similarity for each pair of media items based on underlying media item attributes. Media items are then recommended based on the computed similarities. This approach, however, requires substantial computing resources and is inefficient for vast collections of media items.